nnpd.nre.models
===============

.. py:module:: nnpd.nre.models

.. autoapi-nested-parse::

   Interchangeable classifiers returning logits, not sigmoid probabilities.



Attributes
----------

.. autoapisummary::

   nnpd.nre.models.NETWORKS


Classes
-------

.. autoapisummary::

   nnpd.nre.models.MLP
   nnpd.nre.models.ResidualBlock
   nnpd.nre.models.ResidualMLP


Functions
---------

.. autoapisummary::

   nnpd.nre.models.activation
   nnpd.nre.models.make_mlp
   nnpd.nre.models.make_residual
   nnpd.nre.models.build_network


Module Contents
---------------

.. py:function:: activation(name: str) -> torch.nn.Module

.. py:class:: MLP(inputs: int, hidden: list[int], nonlinearity: str = 'relu', dropout: float = 0)

   Bases: :py:obj:`torch.nn.Module`


   .. py:attribute:: layers


   .. py:method:: forward(inputs: torch.Tensor) -> torch.Tensor


.. py:class:: ResidualBlock(width: int, nonlinearity: str, dropout: float)

   Bases: :py:obj:`torch.nn.Module`


   .. py:attribute:: body


   .. py:attribute:: activation


   .. py:method:: forward(x)


.. py:class:: ResidualMLP(inputs: int, width: int, blocks: int, nonlinearity: str, dropout: float)

   Bases: :py:obj:`torch.nn.Module`


   .. py:attribute:: layers


   .. py:method:: forward(inputs)


.. py:function:: make_mlp(inputs, definition)

.. py:function:: make_residual(inputs, definition)

.. py:data:: NETWORKS

.. py:function:: build_network(inputs: int, definition: dict) -> torch.nn.Module